The United States, China, and India are driving global AI through distinct strategies in policy, capital, and infrastructure. India shows strong momentum but must bridge gaps in talent, data, and compute to secure global AI leadership.
The global artificial intelligence (AI) market is entering a phase of accelerated growth, expected to expand at a compound annual growth rate (CAGR) of approximately 36.6 percent between 2025 and 2030, reaching nearly US$ 1811.75 billion by the end of 2030. This surge also underscores the increasingly transformative role AI plays across domains and industries, from healthcare and finance to manufacturing and governance. The United States (US) continues to lead with the most advanced AI ecosystem, backed by strong research, public and private investments, and policy support. China comes next, driven by large-scale state investments, based on a multi-level framework of laws and compliance requirements. India ranks third globally in AI competitiveness and vibrancy, demonstrating strong momentum, despite gaps in computing capacity, advanced research, development, and innovation.
India ranks third globally in AI competitiveness and vibrancy, demonstrating strong momentum, despite gaps in computing capacity, advanced research, development, and innovation.
China’s rapid rise in AI is driven by sustained and incremental policy support over time. The 2017 New Generation AI Development Plan set clear targets of global AI leadership by 2030, backed by large public and private investments. Public investments such as the US$ 8.2 billion National AI Industry Investment Fund and the broader US$ 138 billion National Venture Capital Guidance Fund for startups and AI-allied fields like robotics are complemented by private investment in AI research and development by large tech companies such as Alibaba and ByteDance. Local governments offer subsidies and relaxed rules in AI pilot zones. Computing vouchers, along with strategic computing infrastructures such as the National Integrated Computing Network, expand access for LLMs. These measures, along with a strong focus on AI education and university-industry collaboration, have created an AI talent pool, enabling quick commercial adoption and a globally competitive AI ecosystem.
China’s AI governance is largely security-first and statute-driven, combining laws such as the Personal Information Protection Law (PIPL) and the Data Security Law with other sectoral rules on algorithms, deep synthesis, and related technologies. These frameworks create obligations such as risk assessments, filings, and watermarking, while giving the state strong control over national security and social order. Through its Digital Silk Road, China is scaling AI by exporting cloud, data centres and ‘safe city’ stacks combined with finance and standards, which can deepen surveillance and vendor lock-ins. Its Global AI Governance Action Plan specifically mentions the “Global South,” calling for international cooperation, joint innovation in AI, and identifying support for “countries, especially the Global South, in developing AI technologies and services” as one of its express objectives.
China’s AI governance is largely security-first and statute-driven, combining laws such as the Personal Information Protection Law (PIPL) and the Data Security Law with other sectoral rules on algorithms, deep synthesis, and related technologies.
This, alongside China’s initiative to establish a World Artificial Intelligence Cooperation Organization (WAICO), proposed to be headquartered in Shanghai, seeks to position China as a global organiser of standards for coordination for developing countries and offer full-stack alternatives to the Global South, despite persisting limits imposed by US technology controls.
The United States emerged as the world’s leading AI hub through a combination of early policy action, significant investment, and strong infrastructure. The 2019 Executive Order on Maintaining American Leadership in Artificial Intelligence and the National AI Initiative Act of 2020 boosted federal research, computing access and AI talent across universities. The Creating Helpful Incentives to Produce Semiconductors (CHIPS) Act strengthened domestic semiconductor capacity, critical for AI.
On governance, priorities are shifting under US President Donald Trump. US AI policy is now moving away from a regulation-based approach toward speed, scale and innovation-led competition. A key aspect of this shift was the rescission of the AI Diffusion Rule in May 2025, which had sought to tightly control the global movement of advanced AI models and high-end chips through a tiered, risk-based export framework. The US now views this rule as overly bureaucratic and hard to enforce, potentially hurting innovation and relations with close allies. Rather than relying on complex diffusion controls, the focus is shifting to securing worldwide adoption of the US AI technology stack and standards through a combination of assertive export controls. This shift is reinforced by the 2025 AI Action Plan, which prioritises deregulation, the acceleration of private-sector innovation and expanding domestic AI infrastructure as a means of strengthening US leadership abroad.
US AI policy is now moving away from a regulation-based approach toward speed, scale and innovation-led competition.
The plan focuses on easing regulatory barriers, supporting not only AI models and scaling chip capacity but also data centres and energy capacity. Internationally, the US is positioning itself as a provider of full-stack AI solutions, combining hardware, software, standards, and security. This approach treats AI as a strategic industrial and security asset, where competitiveness and supremacy are prioritised over safety and guardrails.
India is entering a critical phase in its AI development journey, with the government actively supporting AI innovation, startups, and hubs. The India AI Mission, approved in 2024 with an outlay of INR 10,300 crore over five years, aims to build a shared and subsidised AI compute infrastructure. Furthermore, India is providing access to 38,000 GPUs and 1,050 TPUs, to be offered at highly subsidised rates to enterprises. This seeks to complement domestic chip and semiconductor manufacturing efforts that are already underway, further galvanising India’s AI ecosystem. Efforts like the India Semiconductor Mission seek long-term chip self-reliance, while platforms such as BHASHINI and the upcoming IndiaAI Dataset Platform focus on language access and dataset availability. Indigenous models, Centres of Excellence, and AI skilling initiatives are expanding rapidly, aiming to position India as a future global AI use-case leader. An early indication of the success of these measures may be estimated from major triumphs such as the ‘Mahakumbh 2025’, which witnessed AI-driven crowd management and multilingual assistance through the Bhashini-powered chatbot Kumbh Sah’AI’yak.
India’s model of AI governance is largely characterised as one of ‘Enable, then regulate’, that AI innovation must flourish, but not at the cost of ethics and safety. Accordingly, India’s AI development is supported by the Digital Personal Data Protection (DPDP) Act 2023, along with soft law principles of responsible AI. Recently, India launched its AI Governance Guidelines, outlining seven core principles: trust, people-first development, innovation over restraint, fairness and equity, accountability, understandable by design, safety, resilience, and sustainability. Reflecting Prime Minister Narendra Modi’s vision of ‘AI for All’, the framework aims to balance scale with inclusion and long-term resilience, positioning AI as a tool for inclusive development and broad-based societal benefits.
India’s model of AI governance is largely characterised as one of ‘Enable, then regulate’, that AI innovation must flourish, but not at the cost of ethics and safety.
In operational terms, India’s AI innovation pathway seeks to follow and replicate its pioneering Digital Public Infrastructure (DPI) model, through which the country built a shared public stack of foundational technologies that both public and private innovators subsequently leveraged using open-source software and open APIs to develop diverse applications and services atop this common infrastructure.
Extending this structure to AI, the government is now exploring a similar layered approach, under which core enabling resources such as shared and subsidised computing infrastructure, interoperable platforms, and open public datasets, allowing public institutions, startups, and private firms to build a broad spectrum of AI solutions suited to local needs and contexts. Notably, the DPI model has already generated considerable global interest, with several developing countries adopting and adapting its principles to their own governance and service-delivery contexts. In this light, a DPI-inspired approach to AI development could hold particular relevance for countries in the Global South that are in the early stages of building their AI ecosystems.
India’s ambition to emerge as a global AI hub, however, faces several on-ground structural challenges. A primary challenge is the compute infrastructure. While much has been done to boost semiconductor chip capabilities, the limited availability of energy-efficient data centres, bolstered by high-performance computing facilities and sustainable power systems, inhibits India’s ability to scale AI. It is a vital component in training and deploying advanced AI models at scale.
This issue is anchored in a broader data challenge. Although India has launched notable initiatives such as AIKosh, the Open Government Data Platform, and the National Data and Analytics Platform (NDAP), among others, access to high-quality, updated, and domain-specific datasets remains unevenly accessible. Fragmented availability, interoperability issues, and cautious data-sharing practices reduce the capacity of both public and private datasets to enable high-impact AI use cases.
While much has been done to boost semiconductor chip capabilities, the limited availability of energy-efficient data centres, bolstered by high-performance computing facilities and sustainable power systems, inhibits India’s ability to scale AI.
Another critical barrier is talent. While India produces a large pool of IT graduates, it faces a shortage of advanced AI researchers along with persistent brain drain, limiting its ability to compete globally in frontier AI research. Addressing these challenges requires coordinated solutions, including large-scale public investment in green computing infrastructure, policies that responsibly enable access to datasets and encourage private data-sharing, and the creation of talent pipelines across multiple competencies.
Strengthening academia-industry collaboration, expanding domain-specific research, and incentivising domestic retention of skilled AI professionals are the way forward. These measures can translate India’s AI vision into a sustained, inclusive impact.
Debajyoti Chakravarty is a Research Assistant with the Centre for Digital Societies at the Observer Research Foundation.
The views expressed above belong to the author(s). ORF research and analyses now available on Telegram! Click here to access our curated content — blogs, longforms and interviews.
Debajyoti Chakravarty is a Research Assistant at ORF’s Center for New Economic Diplomacy (CNED) and is based at ORF Kolkata. His work focuses on the use ...
Read More +
PREV
